This episode of The Seller’s Edge features Kiri Masters, a renowned expert in Amazon strategy and digital retail innovation with over a decade of experience helping brands grow. Kiri has worked closely with top sellers and industry leaders to dissect Amazon’s complex algorithms and emerging technologies. Today, she shares her deep insights into Amazon’s groundbreaking AI shopping assistant, Rufus, revealing how this new technology is transforming product search, personalization, and advertising on the platform. Read the full transcript below.
Episode 34 of The Seller’s Edge – Kiri and Jonathan talk about:
- [00:00] Rufus Is Changing Everything
- [01:05] How Does Rufus AI Work?
- [03:56] Are Keywords Obsolete?
- [07:36] What Will Replace Keyword Ranking?
- [09:47] What Impact Will Rufus Have On Ads?
- [12:44] Does Rufus Prioritize Specific Listing Elements?
- [14:45] The Science Behind Click-Training Data
- [15:26] Visual Label Tagging
- [16:33] Keywords In The Image Stack
- [18:20] Leveraging Fringe Use Cases in Customer Reviews
- [21:00] Which Product Categories Will Rufus Impact Most?
- [23:46] How Will Rufus AI Evolve In 6-12 Months?
- [25:04] Open AI Operator’s Retail Capability
- [26:06] The Way Loyalty Programs May Change
- [29:25] Has This Altered Your Approach To Retail Media?
- [32:34] What Does This All Mean For The Future?
- [37:42] Recap and Closing Remarks
Key Takeaways:
- Focus on clear product descriptions over keyword stuffing
- Optimize for engagement and conversions, not just search rank
- Use customer reviews to discover and highlight new use cases
- Prepare for AI-driven personalization in search results
- Stay agile and adapt quickly as the landscape evolves
- Invest in clear brand differentiation to stand out
Full Transcript of Episode:
JONATHAN: For the sellers out there who haven’t experienced it. Can you explain how Rufus interacts with shoppers and how that’s different than the traditional Amazon search?
KIRI MASTERS: Yeah, absolutely. So we’re at a, we’re at a point in time where we’re seeing like a little bit of a pivot and I really believe this is my, and I’ve got, I’ve gone on record on this and I’m happy for people to call me out in two years. But I think in two years we’re not going to be searching using like the traditional search bar with keywords as our primary mode of interacting with a retailer. And the reason is I looked to Rufus and what’s happening and there’s some researchers that found the patent and dissected it and we’ll get into more of the details on that in the interview. But Rufus is an AI shopping assistant that’s more conversational and it’s surfacing products in a way that is more natural and you know, less about sort of gaming the system with keyword stuffing and things like that. And it can be a little bit of a frustrating experience for shoppers to go to Amazon and just have you know, these be surface, the same, all the same products and like I don’t know which one’s for me and why is this one priced differently to that one. And Rufus as a conversational AI agent is able to discern the difference between different products better and understand what you’re really asking for as well. Like dig into, you know, you’re searching for. Sometimes there’s like a problem based query that we have like you know how, how the example in this report was how to remove gel nails. So Rufus understands what type of nail polish remover you need based on that query rather than you having to use the actual search term that you need for that product. So it is a more intuitive way of searching for products. It is more, it’s more helpful. Rufus is showing that it personalizes search results as well. So it might be based on your past shopping activity. It might understand things about your nate, your lifestyle or your sort of financial thresholds, whether you’re a value shopper or you tend to splurge a little bit. And so I think that we’re at this point of an inflection point where Amazon has come out with this model. We understand a little bit about how it works now and Amazon’s really the canary in the coal mine for retail. And once we see Amazon Adopting some form of technology, we often see other retailers following suit.
JONATHAN: Yeah, for sure. That’s interesting. So it’s more. Because I’m used to, I mean I started in Google SEO and that’s really more information based questions and it’s, it seems like it’s more aligning to that than it is transactional. You know best nail, polish remover or whatever you would be searching for. That’s interesting. I mean keywords, everyone’s been singing the death knell of keywords for quite some time. It’s interesting because last year Amazon expanded the backend keyword field and it was like 250 characters and then it was like 2,500 characters. And I was like why are they doing this? Because it seems like keywords don’t matter anymore. So it was just a very bizarre thing for Amazon to be doing. Especially with the advent Rufus. I mean who knows why Amazon does the things that they do.
KIRI MASTERS: Yeah, that’s very interesting because it certainly is a trend in the other direction like you said. But Amazon is such, you know, such a huge company now and it’s very possible that you know the left hand doesn’t know what the right hand is doing essentially. And we’re still at a point where Rufus is I was actually seeing keyword research. I was doing some, some like some SEO research the other day. I was looking at what are people searching for about Rufus? And one of the top queries was how do I turn Rufus off? So, so I don’t think that consumer adoption is going to be a little slow. People kind of fight it. Especially initially. There’s a lot of dismissing of Rufus and how it wasn’t that intelligent and I wrote it, wrote an article last year, found a, a thread on Reddit where someone had asked Rufus, why are you named Rufus? And Rufus said I’m not named Rufus. So I don’t know. They didn’t know what its own name was. So I think that this type of thing sort of like looking at the whole AI landscape overall, which I know we’re going to talk about and looking at the first iteration of a tool or a technology like ChatGPT when it first came out GPT3 or whatever it was and look, you know, try out a couple of use case and say well it can’t write right very well, it’s not going to take over my job. Like this is this is not that good. But we don’t understand like it, it’s hard for us to comp, comprehend how much it will get better and how quickly it will get better. And the, you know, the, the first version of something out of the gate, technology wise isn’t often like that. Great. It’s easy to poke holes in it. I think that that’s what a lot of people are doing, which with Rufus, with ChatGPT chat, sorry, OpenAI operator and looking and saying, well, it can’t do this, this and this. And this part is clunky, so this is never going to work. And I think that that’s a, recipe for being proven wrong in the long term.
JONATHAN: Yeah, for sure. I think AI, I mean, you said it, it’s in its infancy and we really, I mean it’s been tested and qaed behind closed doors, but you don’t really know what it’s going to do until you have the entire world trying to battle, figure out how, what it’s, you know, limits are. So I think that the only way to do that is actually start and that’s probably where we are right now. I don’t know why Amazon does any of the things they do as far as the backend keyword search terms. I mean that could have been something that they were doing just to test Rufus in some way. So they wanted to like have everyone put a bunch of backend keywords out there. But I’m curious. You mentioned the personalization piece and it’s funny because I’ll use a tool like Viral Launch and I’ll look up a product listing and I’ll say, you know, it’s ranked, you know. Cause it’s almost like true rank is like number eight in organic. And then I’ll log on Amazon in an incognito window and I’ll do a search term just to see if that’s where it’s ranked. But it seems like that wouldn’t be the case anymore because I feel like everything is going to be so personalized. Not that it’s not already, but I just feel like to that extent it’s going to be harder, harder for us to kind of gauge where everything falls.
KIRI MASTERS: Yeah, I wonder about, I wonder about search, rank and what relevance that would have. I think that Rufus is. You think about Amazon’s overall objective, and this is the same as it’s always been, is to surface relevant products for that customer that they’re going to buy. And that’s the whole point of, of of, you know, BSR and all of the different aspects of the A9 algorithm that we, that we know is that Amazon wants a transaction to happen and a transaction is most likely to happen on a product that has proven that it’s already sold very well. A product that already has like really great ratings and reviews a product that it’s a known quantity and it’s you know, value priced and things like that. So I think Rufus is still going to take a lot of those factors into consideration and if someone’s searching for a given term it’s going, it’s going to want to surf all other things being equal, it’s going to want to surface a product that has a proven sales history and has proven to, to be well rated and not have high returns and things like that. Because Amazon wants a sale to happen and for that custom to, to be, to have found what they were looking for. So that’s a really good. That didn’t really come up in the the Rufus AI patent as a, as a factor the the sales rank numbers. But yeah it’s interesting to think about what kind of role that would have.
JONATHAN: I don’t think anyone has the, all the answers regarding Rufus including Amazon right now. So I think it’s largely like trying to figure it out because I’m curious about things like that. I’m also curious about like you know when it’s looking for a sure fire product like what does that mean for new sellers who are just getting their listings onto the marketplace and like can they pay enough to get it to the top of Rufus’s recommendations? If it’s right, it’s not there already. So it’s like how do we get that foothold for new sellers?
KIRI MASTERS: Yeah, that’s a good point. So there’s a couple of things I can add to here. So one of the, and I do recommend people check out this report. I’ll make sure that you have the link for it. But it was looking at the Rufus AI patent and explaining what it means. One of the elements of the patent is what Amazon calls and they have to name all of this stuff. So it’s interesting to hear what they name these these principles. But one is called Qlik training data. And so that is to what we kind of talked about a little bit. It’s create, it’s looking and learning what products customers actually click on after specific queries. So that Qlik training data is learned is, is learning user behavior particularly which products people click after queries and that creates a feedback loop. So that actually might be sort of a factor beyond what the search you know what your product rank is There and then the other aspect is advertising. So in a really like a developer release notes from I think like November last year, Amazon did say something about the potential for Rufus to have ads within. Like they were building out a capability for Rufus to potentially have ads in, in the, in its system. So I think ads, look, ads are not going to go away guys. This is, this is a major, major income stream for Amazon. That is one thing we do not need to worry about is paying to get your products in front of people. It might look a little, we’re not going to see. You know, it won’t look, it’ll look a little bit differently when it’s within Rufus to when it’s on a search results page. But yeah, we do not need to worry about there being no ads on Amazon in the future. That is not going to happen.
JONATHAN: Yeah. And then like, I mean as far as Rufus analyzing product listings because I mean those aren’t going away either. I mean is there a certain element in the listing that Rufus might prioritize other like over others? I know people right now like the title is always important but is, is Rufus looking at use cases in the bullet points or project description over other things?
KIRI MASTERS: Yeah. So this is where I as a consumer and this is why I think it will. One of the reasons why it will stick is I come across these products listings sometimes that are so keyword stuff. I don’t know what the product is. Right. I’m sure everyone listening has seen that. And you’re like I will look at a dress and it will say midi skirt, maxi dress, tight, flowing. I’m like what is it? Like I can now I have to look at the image and I have to kind of like take a chance on this product which is like just been so over engineered that I don’t know what it is and I don’t know if I’m going to like it and it’s confusing. So keyword stuffing is not going to be helpful in the future. And I think that it is a good thing for consumers. So the question is, well what is Rufus going to be looking for? It will be looking at your product listing for content but it’s going to be written content like the description, the keywords, but also things like the product attributes as well are going to be really important. So it might be less about like I need to sell this product to a consumer when they land on the page and a little more importance might be around really accurately describing that product and the use cases so that Rufus Understands it. And this is something we’re going to see sort of across, across the board in consumer marketing, whether we’re talking about Amazon or we’re talking about other retailers, is optimizing for an AI agent as well as the front end consumer. So yes, that’s going to be the context of how this product is used. Its attributes, dimensions, nutritional information, things like that, which you know, probably already including to some degree in a product listing. But you think about how like, how is Rufus going to explain this to someone? The other element is this like sort of like three foundational pillars of Rufus. The other element is that click training data that I mentioned where that’s almost like its own bsr, I guess, like it’s going to look at user behavior and which products people click and purchase. And I think that that has the highest potential for personalization as well. Like looking at you as a consumer and the fact that you have like. Do you have children at home? Yes or no? Well, we based on past interactions with, with users, like the people with children at home tend to select this one versus that one for people without children at home, for example. So that’s click training data. And then the final part of the final pillar here is visual label tagging. So this is very interesting. We’re not just looking at like text based content for indexing with Rufus. We’re also looking at images. So this is a big part of the pattern where diagrams, how to kind of content, how how something looks, its physical attributes. Those are all being processed by Rufus. I guess in some categories it matters more than others. But we’re also, you know, we’re going to see Rufus documenting what is happening in images. So that is important to think about as well with like know your diagrams and use cases as well. So this might come down to the product. The lifestyle imagery that we put with products for example is less about selling like this lifestyle to a customer and could actually be telling Rufus, hey, this is, this can be used outdoors, this can be, you know, what, what are you, what are you telling the, the AI about your product with each image as well.
JONATHAN: Yeah. So I would imagine it’s also probably going to be useful not to keyword stuff there because I feel like the next iteration of what people are going to is just cramming pictures with all sorts of specs and information, thinking that they can rig the system with that as well.
KIRI MASTERS: It’s a great point. And my hope for Rufus is that that behavior won’t be rewarded because that’s Just not a good customer experience, is it? So what I, what I hope would happen in that scenario is the products that are like so over engineered and stuffed with information. Rufus will look at that and say I don’t know what this is, who it’s for or how it’s used. So I’m probably, I’m not going to surface it very often. Whereas a product which has, let’s say like a product has use cases for different types of people and that’s fine. So it can be used for moms and for you know, the elderly and it talks about those in the product description and in the images and stuff, that’s okay. But if it’s like this is for everyone, it can be used at every, you know, occasion and I think reevaluate is going to be that. Well that’s not really possible. I don’t really know who it’s for or what it does or when to use it. So it’s, I’m not going to surface it as much as a product which is, here’s the use case. It can be used for, you know, these, you know, three or four different things. Here’s who it’s suited for. That is, that’s clear to the AI, just like it would be clear to a person as well. Like this is, is this for me or is this not what I’m looking for and I can, I can move on.
JONATHAN: Yeah. I’m curious, do you think that it’s also because, I mean Amazon now uses the word cloud that finds the most commonly used phrases in reviews. I’m wondering if that also factors in, if they’re using, looking for use cases that might be mentioned in reviews that don’t necessarily come up in the product listing. Like I find products all the time where I’m like it’s so interesting that people use this for, I don’t know, a doorstop or something when it’s like a kitchen blender, like something else. Yeah, yeah. So I would imagine that that would probably come into play too. Right? Or I mean if you had to guess.
KIRI MASTERS: That is a really, really great point. I think 100%. I think when I’ve been playing around with Rufus it will call out attributes that reviewers have mentioned and that’s really helpful. Shopper. I want to know what other people have used it for, what other people think. So it’s totally, it makes total sense. Give you an example, I haven’t searched for this but I think it will be I’ve got a hunch so There’s a, there’s a trend. Is or was, I don’t know, still happening. There was this trend in beauty routines where people were using diaper cream, in their beauty regime. And it’s like something about that, I think it’s zinc or something in the active ingredients. It’s like quite good for your skin. And so people were like, buying, diaper cream for. Not for babies, but to put on their face at nighttime. And that is not going to be, generally speaking, a product feature that you would put on your diaper cream product listing. Right. It would just be. It’s. It’s sort of an edge case and it’s going to confuse people. Like, yeah, I don’t need to say anymore, but it’s going to show up in the product, reviews people will talk about. I’m using it for my nighttime skin routine and stuff like that. And so that’s where I think, you know, rufus and other AIs will. Will be able to like, find those use cases that people are mentioning in product reviews, but not necessarily in the description and surface those as well. So people are talking about, they called it, they called it something ridiculous. It was. They called it face basting. You learn something new every day. I feel, I feel like I’m teaching you a lot here.
JONATHAN: You are. And I feel like in. In the age of TikTok. In the age of TikTok, like, things like that are just more and more common because there’s like, not a day goes by where I don’t find some strange video where someone’s like, using a kitchen spatula to change the oil on their car.
KIRI MASTERS: Yes. Exactly.
JONATHAN: You mentioned product categories and I’m curious, do you think that Rufus will have a greater impact on like, one product category versus another? And if so, how do you think sellers should sort of be adjusting their strategies for that?
KIRI MASTERS: I will say I don’t think any category is really immune to this. Lots of, lots of people in the, like, apparel and fashion categories, they’re the first ones to push back on, on AI. I’ve found when I’m, when I’m talking about this and it’s not just me, but, the pushback is. Well, people are always going to want to try on clothes in, in store and we’re never going to stop going to stores. People want to try on their jeans before they buy them. Fashion is this expression, you know, self expression and all this kind of stuff. And those things can be true. Are true. I Like going to stores and trying things on as well. But it’s not every use case either. And I think something that really snuck up on the apparel category is she in. And Shein is really like the. Has disrupted fashion through AI So Shein is looking at what, what looks are trending on social media and they are using AI to order in, you know, these styles of different factories and move very, very quickly from, you know, what’s, what’s a trending look on TikTok or Instagram or whatever, and moving that to that item to being on for sale on the app in two weeks. So you look at all, you know, the huge volume of sales going through Sheen for fashion, and you cannot tell me that people only shop in stores where they can try stuff on anymore. Like it. It is so, so popular. For better or worse. That whole assumption that fashion can only happen in stores has been completely blown out of the water. I don’t think we’re, we’re going back there. There’s of course, like some. This is the things like we can’t talk in absolutes here. In some use cases, I want to go and try things on. In other use cases, I’m looking for a cheap thrill. In other use cases, I’m looking. I know what size jeans I am and what brand I, like I’m going to reorder the same pair of jeans. So I think it’s unwise to say like, any category is immune to it because history has just shown that not to be the case Sellers, every category are going to need to get ready for this.
JONATHAN: Yeah. It’s funny just because I think the trend that I know of, of people that I’m in my. In like just people in friends and family circles is just ordering seven different versions of something offline for apparel, trying them on, and then returning the rest, which sellers don’t like, but consumers do. And then for Rufus, I mean, it’s just crazy to think of where it’s at now. And again we mentioned it being in its infancy and just knowing where or not knowing where it could possibly go. And I’m curious, do you have any thoughts as what we might see as capabilities that come out in the next six to 12 months? Like, what does that look like?
KIRI MASTERS: Yeah, next six to 12 months. Look, I think that, maybe shifting gears a little bit. I think, you know, Amazon has, have, a. Has a very innovative concept with Rufus, and I think they’re very, very much further ahead than many, if not all other retailers at this point in time. With that having their own shopping assistant at the same time. We just saw OpenAI launch operator. Well like two weeks, a week and a half, two weeks ago as, as of at the time we’re having this conversation and I think that that has the potential to disrupt a little bit that the whole, the whole retail sort of category and the way that we, we’re shopping. And again this going back to, to what we talked about at the beginning of the episode, suspending your disbelief a little bit that you know, the current iteration of what we have is not going to get any better. So just to, to recap, anyone that didn’t sort of catch this as it first came out probably heard about it. But what the, what the retail implication for Open AI Operator is? I can ask, I can tell Operator. I’m going to make these three meals this week, figure out what the recipe, what the, what the shopping list needs to be and then go and like go and do a price comparison at like these three retailers and just put the order in at the retailer that has the best total price and, and have it delivered on Tuesday at 9am To my home. And then AI is going to go and do that for. He’s going to go like, figure out what do you need to buy. Maybe consider like what are your preferred brands as well. Like you know, house brands are okay for these categories but Carrie, you know, she has to have like this certain brand of cereal. She doesn’t want to have a replacement and go and do a price comparison at a number of different grocery stores and place an order with the one that has the best total price or whatever dimension I’m looking at. Maybe I think this is where like the loyalty programs are really going to come back because my decision to shop at a certain retailer might be swayed by accumulating points or getting like better, you know, getting a fuel discount or something like that. So this is where things are going and I think it’s going to get to that point very, very quickly. This raises lots of questions around what is the, who has the relationship here? Is it the retailer or is it the AI agent in this case? And now there is this like intermediary between the consumer and a retailer. That is potentially where we’re going to. If you’re a brand, you’re going to want to place ads. Because I’m not going to be going to, you know, kroger.com anymore to do my, to, to, to place my shopping order. I’m just going to be asking my agent to go There. So I’m not going to see those ads and an agent’s not going to care about ads. So it’s very. This is throwing this door wide open on the job of a retailer and is this going to necessarily, is this going to be like the retailer just becomes a fulfillment, the fulfillment part of the chain? I don’t think so. I think we’re going to find ways. Retailers are going to find ways, like I mentioned with loyalty programs and things like that to be relevant. Amazon has, you know, in the case of Amazon, Amazon has a huge advantage in this. The, the breadth of its assortment. Like you can just find anything you could want at Amazon. Right. And the prime membership is a really strong program as well. People know that something’s going to show reliably, it’s going to show up quickly and I can return it really easily. So people might still choose Amazon for those products even if they’re using this like AI agent on the front end. So I think long term that’s where we’re going to be headed is a lot more. We’re going to use agents like either our own personal agents or sort of these. I’m not sure if you’d call them agents or assistants on the retailer.com websites, including Amazon. That’s the long term view for me. Like that’s a, that’s sort of like a two year time horizon, six month time horizon. A little harder to, a little harder to, to project out for me. I don’t think like, I think it’s just going to kind of. Rufus is just going to be a little bit less annoying and a little bit more helpful in six months. That’s, that’s really the only prediction I can say. Like maybe we’ll see less of those queries about how to turn it off. People might use it again. This is the challenge with like shipping something so early is that you ship something that’s not, you know, super helpful out and people make a judgment call pretty quickly. This is not helpful. I don’t like it. I’m not going to use it. I want to turn it off. So I think the next thing is for Amazon to really prove how useful it can be. That will be the next step.
JONATHAN: Yeah, we are officially living in the future if, if anyone has missed it. It’s so interesting just because, yeah, like the AI assistants because you mentioned, like would we be using kind of like a third party or the retailers ones. And I’m just, I mean I really feel like that’s just going to depend on consumer behavior and what they end up using. Because that’s going to be the thing I’m curious from like a, like a retail media perspective, like how do you, like how do you even begin to approach that? Or like has that already shifted your thinking or your approach to retail media in general?
KIRI MASTERS: Oh, this is, it’s, it’s so interesting to think about what could happen here because like I said, if we don’t have consumers interacting directly with the retailer website or app and it’s, you know, it’s agents interacting with a retailer on the advertising side. Yeah, the agents are not going to be looking at sponsored products, listings are going to be, they’re going to be going directly to, to purchase a product. But what I see happening is the AI agents, like a personal AI shopping agent is going to understand my preferences for like price, brands, categories, things like that. And some of this retail media ad spend that we’re expecting to hit retail retailer PNLS in the coming years is going to be reallocated to AI agents because that’s where the consumer interface is. And whatever sort of layer or surface the consumer is interacting on, whether that’s in store or at a.com or on Instacart, as Instacart is another intermediary as well, that’s where the ads are going to go. They’re going to go to the, to the, the point where the consumer is interacting. So I think that we’re, we will see these AI shopping agents sort of monetized through ads and that will be those, those can be like helpful promotions as well. Like hey Kira, I know that you you’re going on vacation in a couple of weeks and I noticed that your favorite sunscreen is on sale. Do you want to stock up now from Amazon? Things like that, where it’s like it understands my purchase behavior might put new things in front of me as, as a result of, of that I think that more ad spend is going to be spent on brand type advertising, display and brand advertising and it will be more on these sort of partnerships that retailers have with media companies. So you know, Amazon has Prime Video so it has its whole own ecosystem of media that it can sell advertising on. But other, other retailers have partnerships with Roku and Paramount plus and other networks. So I think that the advertising is going to be sort of potentially less about these sponsored product ad placements and more about brands on those on those media properties that would, we won’t stop watching TV in the future, but we might stop going to kroger.com to do our own shopping.
JONATHAN: Like what kind of business models and partnerships does that all create? Like I’m even beginning to just like grasp at straws thinking about like, well this is kind of, we’re so used to the retail experience as we’ve known it for a long time. Amazon has just been sort of a digital version of that. But this is a whole other, this is a horse of another color. So how do we, what does that even begin to look like for the future of retail and partnerships and business models in this space?
KIRI MASTERS: Yeah, it’s a, it’s a big question and I think that look, that the sell, the, the third party seller, ecosystem on Amazon has changed dramatically from like, you know, I got started in this space almost 10 years ago now and it’s been so many evolutions and I think the people listening to this show who have an Amazon business, you’ve had to deal with a lot of like, changes and disruptions and like things have gotten more difficult in terms of fees and now you have to spend more money on advertising. And I think there’s, there’s a survivorship. The survivorship of this type of business is, is, is pretty important. So I think if you’ve gotten this far, you’re going to find a way to figure it out. Right. Like we were talking about at the top of the episode that there’s a lot of this is saying like, show me, tell me the incentives and I’ll tell you the behavior. Like keyword stuffing was a tactic for a reason. Like it, it meant that your product would show up in more search results for different phrases. And that’s not going to work in the future. But there’s going to be other, there’s going to be other elements where we’ll figure out, oh, you know, by having this certain, type of image. Rufus really likes to have comparison tables or something like that. Well then every, every image has a comparison. Every product has a comparison table on it within. So I think that just as we’re seeing this, this, the, the ground sort of shift under our feet and we’re not sure exactly, you know, what does this mean for product optimization? What does this mean for ads? What’s going to work in the future? We will figure it out just like we always have and there will be opportunities. If you’re tuned in on this kind of stuff and you can kind of see, okay, what is the net, I don’t know what things are going to look like, what things will look like exactly in five years but what is the next thing that I need to be ready for and what do I need to optimize for next? That’s the best practice. So I think, you know, people who are listening to this show, are interested in staying in touch with what’s going on and skilling up and what’s the next edge that I can get. And that’s really where it’s, where it’s all going to happen. It’s just going to be sort of one step at a time. What’s working now, what’s working next, what do I need to think about next? And to be able to keep moving forward.
JONATHAN: Yeah. And in case any of what I said came off as like a fear or doubt or skepticism, I’m excited about all of this. It’s just like the possibilities are endless.
KIRI MASTERS: Yeah, exactly. And I think, you know, what we’re talking about is it’s not going to just like happen next week. It’s going to be this evolution over time and we will see, there will be opportunities for. Look, people, the whole reason Amazon is popular and Shein and Temu, as much as, you know, those can be competitive challenges for sellers. The point is people want choice. They want choice around features, price, all these different dimensions. And ultimately with Rufus, for example, Rufus is going to like giving people multiple choices because that’s what people want. They may not want 200 different options that all look the same. That’s not good for the consumer and it’s not really great for the sellers either. But if there is discernible differences between product A, B and C, and Rufus can say, product A is great for this reason, product B is a bit more expensive, but you get this with it. And, product C is like, you know, the most, you know, has the best reviews. That’s a great customer experience. And I think that the sellers and the brands who, can work into that scenario and not just be copycats, but have discernible features and benefits that a certain group of customers are looking for, that’s going to be the same as it ever was. People want choice. They want to feel like they made the best decision and those, those things are never going to go away.
JONATHAN: Absolutely. Extremely well put. Kiri, you have been a really amazing guest. You’re very knowledgeable. I appreciate you coming on and sharing all of this, with us. I mean, there’s just so many other kind of cans of worms that I feel like we’ve opened in the conversation. I’m like, I want to have a conversation with her about that. So at some point, we’ll just have to have you back so I can ask those additional questions.
KIRI MASTERS: Great. Well, thanks for having me.
